NEWS.md
aggregate_impute()
handles the corner case when join
results in an empty dataset.model_fun
argument of model_impute()
can be either a function or a string containing the name of a function (like "glm"
). Include the package name in case the function is not available in base R (like "INLA::inla"
).impute()
gains an extra
argument. Use it for observations not in the model that you still want to add in the follow-up analysis. For example: exclude rare observations from the model but you want them in the aggregations.impute()
on INLA models now also handles the binomial, the zero-inflated Poison (type 0 and 1) and the zero-inflated negative binomial (type 0 and 1) distributions.hurdle_impute()
to fit a hurdle model based on a model of the presences and a model of the counts.rawImputed
and aggregatedImputed
objects.checklist
infrastructure.checklist
infrastructure.aggregate_impute()
now also works on aggregatedImputed
objects (#34)